Codex Sana’a and the Qur’an
Bismillahir Rahmanir Raheem
بِسۡمِ ٱللهِ ٱلرَّحۡمَـٰنِ ٱلرَّحِيمِ ,
There have been numerous articles, books, lectures given on the topic of the Sana’a MSS (manuscripts), but they always seem to fall short of explaining the reality of the manuscripts. What does the existence of the manuscripts mean for the Muslim and Christian community? Does it truly prove that the Qur’an has been ‘corrupted’? How reliable are the findings? Insha Allaah (God willing), I seek to answer a few of these questions and more, while aiding in the understanding of the manuscript tradition (Ulum al Qur’an, Textual Criticism) in light of the Sana’a codex.
An Introduction to the Sana’a Codex:
A codex is simply a ‘collection‘, so the Sana’a codex is the ‘collection of manuscripts from Sana’a, Yemen‘. The Sana’a manuscripts were discovered accidentally, the event is actually, quite famous history:
The manuscripts, thought to be the oldest surviving copies of the Koran, were discovered in the ancient Great Mosque of Sa’na in 1972, when the building was being restored after heavy rainfall, hidden in the loft in a bundle of old parchment and paper documents. They were nearly thrown away by the builders, but were spotted by Qadhi Isma’il al-Akwa, then president of the Yemeni Antiquities Authority, who saw their importance and sought international assistance to preserve and examine them.
From thereon, the services of Dr. Gerd Puin had been consulted:
Al-Akwa managed to interest Puin, who was visiting Yemen for research purposes in 1979. Puin in turn persuaded the German government to organise and fund a restoration project. The restoration revealed that some of the parchment pages dated from the seventh and eighth centuries, the crucial first two centuries of Islam, from which very few manuscripts have survived.
Dr. Gerd Puin’s Findings and Conclusions:
Puin noticed minor textual variations, unconventional ordering of the chapters (surahs), as well as rare styles of orthography. Then he noticed that the sheets were palimpsests – manuscripts with versions written even earlier that had been washed off or erased. These findings led Dr Puin to assert that the Koran had undergone a textual evolution. In other words, the copy of the Koran that we have is not the one believed to have been revealed to the prophet. – 
Understanding the Findings and Conclusions:
It is far too easy to condemn (if you’re Muslim) or support (if you’re non-Muslim) something when it’s aiding or detracting from your cause/ belief. For the Muslim, these findings are supposed to prove the Qur’an is corrupted and thus their word of God is false, to the Christian missionary, this is finally their opportunity after many failed attempts to prove that the Qur’an is not the original word of God, an argument made popular by Muslims against Christendom for several centuries. Yet, do the findings really reflect these views? For the learned, we’d answer with an emphatic “no“. Let’s examine what the findings state and then put them into context:
- Minor textual variations.
- Unconventional orderings of Chatpers/ Surahs.
- Rare styles of Orthography.
Yet, before we examine the three main points above, there is a key argument that we’ve ignored, and now I’d like to bring it to light. The topic of ‘palimpsests‘.
Palimpsests and the Qur’an:
A palimpsest can be defined as, “a parchment or the like from which writing has been partially or completely erased to make room for another text. – ”, usually this is the missionary’s main argument when using the historicity of the Sana’a codex. They assert that because the manuscripts show that there was a ‘previous text’ before which was removed and then the current text written, the conclusion has to be that errors were made, and then emendations (alterations to text for improvement) were made, hence the current version of the Qur’an is an update from the imperfect older version as is seen in the Answering Islam article, “The End of the Qur’an as Muslims Know It“, and Faith Freedom’s, “Ancient Qur’anic Manuscripts of Sana’a and Divine Downfall“.
However, this argument can only be blamed on abject desperation and a severe lack of education in the field of textual criticism. The world, for a vast amount of its history relied upon liturgical (oral – speaking and aural – listening) transmission of data and information. The use of texts as a primary form of transmitting data, that is, textual transmission did not become standard in the late 15th century with the advent of the printing press:
Printing with movable type had existed in East Asia at least since 1377 when the Jikji, an abbreviated title of a Korean Buddhist document was printed in Korea during the Goryeo Dynasty, however, the invention had not spread to Europe where everything people read still had to be copied by hand or printed from wood blocks carved by hand. In about 1440, the German goldsmith, Johannes Gutenberg, developed a movable type. Gutenberg made separate pieces of metal type for each character to be printed. With movable type, a printer could quickly make many copies of a book. The same pieces of type could be used again and again, to print many different books.
Printing soon became the first means of mass communication. It put more knowledge in the hands of more people faster and more cheaply than ever before. As a result, reading and writing spread widely and rapidly. – 
Similarly, this author goes into a bit more detail as to the development of texts (books) as a primary form of communication and mass media:
Nonetheless, books were hardly considered a mass medium because very few copies existed. Until the middle of the fifteenth century, most books were hand-copied, often times my monks. Such books were expensive and very few people could read or write. As a result, only religious orders, the ruling elite, and some wealthy merchants ever saw or owned one.
Probably the most important milestone in the development of mass communication came in 1456 with the invention of the printing press and movable type. In Mainz, Germany, Johannes Gutenberg paved the way for the reproduction of books for the masses … – 
Thus the use and reuse of manuscripts was common, as Dr. Bruce Metzger and Dr. Bart Ehrman indicate with this following excerpt:
In times of economic depression, when the cost of vellum increased, the parchment of an older manuscript would be used over again. The original writing was scraped and washed off, the surface resmoothed, and the new literary material written on the salvaged material. Such a book was called a palimpsest (which means “rescraped,”). One of the half-dozen or so most important parchment manuscripts of the New Testament is such a palimpsest; its name is Codex Ephraemi rescriptus. Written in the fifth century, it was erased in the twelfth century and many of the sheets rewritten with the text of a Greek translation of treatises or sermons by St. Ephraem, a Syrian Church father of the fourth century. By applying certain chemical reagents and using an ultraviolet-ray lamp, scholars have been able to read much of the almost obliterated underwriting, although the task of deciphering it is most trying to the eyes. In A.D. 692, the Council of Trullo (also called the Quinisext Council) issued a canon (no. 68) condemning the practice of using parchment from manuscripts of the Scriptures for other purposes. Despite the canon and the penalty of excommunication for one year, the practice must have continued, for of the 310 majuscule manuscripts of the New Testament known today, 68 are palimpsests. – 
As I stated previously, only someone ignorant of manuscripts and their study can assert the claim that palimpsests means that a text is invalid, corrupted and emendated. While this can be the case, and I’m not saying it can’t, it is more likely that due to the cost and availability of fresh writing material, many opted to wash over and rewrite on the same manuscripts. If we take the missionary argument that the existence of a palimpsest proves the corruption of the Qur’an, then the Bible must be overwhelmingly corrupted as one major manuscript and 68 others (as documented above) fall pray to this practise. Clearly the beat the missionary is marching to will come to an abrupt halt with such information.
Palimpsests: Textual Variants and Orthography of the Qur’an:
Palimpsests for the Qur’an are extremely rare, with only one other codex known to contain them . The scriptio inferior text, is the text that was washed and then written over, in the case of the Sana’a manuscripts, the scriptio inferior were never the same as the scriptio superior (the newly written text) – . This would have to lead one to belief that the manuscripts were washed and rewritten to accommodate a rewriting of the texts for preservation (verses) as is testament by the Sothbey 1993 and Stanford 2007 folios where the original writing contained 30 verses, spanning 2:191-223, after the washing, the verses found on that same manuscript were lessened to 20 verses, from 2:265 to 2:286, thus accommodating an overall rewriting of the texts for preservation.
|After Wash||Before Wash||Source:|
|2:265 – 2:271||2:191 – 2:196||Sotheby’s 1993 / Stanford 2007, recto|
|2:271 – 2:277||2:197 – 2:205||Sotheby’s 1993 / Stanford 2007, verso|
|2:277 – 2:282||2:206 – 2:217||Sotheby’s 1993 / Stanford 2007, recto|
|2:282 – 2:286||2:217 – 2:223||Sotheby’s 1993 / Stanford 2007, verso|
Some may question why there was a rewriting of the text, and this is a fair but common question. Texts are usually rewritten to reflect the standardization of a language, as is testament also in the Greek language (orthographical development, i.e liturgical transcribing):
Ancient scribes, when writing Greek, ordinarily left no spaces between words or sentences (this kind of writing is called scriptio continua), and until about the eighth century punctuation was used only sporadically. At times, of course, the meaning of a sentence would be ambiguous because the division into words was uncertain. In English, for example, GODISNOWHERE will be read with totally different meanings by an atheist and by a theist (“God is nowhere” and “God is now here”).
Furthermore, it should not be supposed that scriptio continua presented exceptional difficulties in reading, for apparently it was customary in antiquity to read aloud, even when one was alone. Thus, despite the absence of spaces between words, by pronouncing to oneself what was read, syllable by syllable, one soon became used to reading scriptio continua. - 
Similarly in the Arabic language it was common for one tribe or city to speak one way, but write another, yet both having the same pronunciation:
Some tribes would pronounce the word حتی (hatta) as عتی (‘atta), and صراط (sirat) as سراط (sirat), etc., and this was the root cause of many of the known variants in recitation. Similarly the letters ا, و, ي have the dual function of consonant and vowel, as in Latin. The question of how early Arab writers and copyists used these three letters requires special attention. Their methods, though puzzling to us now, were straightforward enough to them. – 
Something which we can also identify with in the English language:
The Boy of Bilson: or, A True Discovery of the late notorious Impostures of certaine Romish Priests in their pretended Exorcisme, or expulsion of the Divell out of a young boy, named William Perry, sonne of Thomas Perry of Bilson,in the country of Stafford,Yeoman. Upon which occasion, hereunto is permitted A briefe Theological Discourse, by way of Caution, for the more easie discerning of such Romish spirits; and iudging of their false pretences, both in this and the like Practices. – 
The following words may have seemed odd to you, but they were the earliest transcribing of the orthographical development of the English language:
- certaine – certain
- exorcisme – exorcism
- divell – devil
- sonne – son
- briefe – brief
- easie – easy
- iudging – judging
No one would dare say that rewriting certaine as certain is a corruption of the text, nor would rewriting divell as devil be considered corruption. These however are attempts at preserving a text as the language itself becomes standardized. As it would seem, the Sana’a manuscripts record the orthographical standardization of the transcribing of the Arabic language. In other words, the Qur’an has not been altered or corrupted, but preserved in its original tongue and to do so would have been to preserve the text by transcribing it accurately. This understanding can also be found in the hadith of Anas bin Malik (may Allaah be pleased with him) who has narrated, “‘Uthman [ra] then ordered Zaid bin Thabit, ‘Abdullah bin AzZubair, Said bin Al-As and ‘AbdurRahman bin Harith bin Hisham [raa] to rewrite the manuscripts in perfect copies. ‘Uthman said to the three Quraishi men, “In case you disagree with Zaid bin Thabit on any point in the Qur’an, then write it in the dialect of Quraish, the Qur’an was revealed in their tongue.” They did so, and when they had written many copies, ‘Uthman returned the original manuscripts to Hafsa.” 
The Qur’an (literally: The Recitation) when being transcribed was transcribed according to the vernacular of some scribes from throughout Arabia, therefore Uthman [ra], ordered the scribes to write the Qur’an by transcribing it, according to the tongue (liturgical transmission) of its original revelation: the Qurayshi Dialect. Therefore any such palimpsest of the Qur’an which demonstrates orthographical variants or textual variants is due to the Arabic of the Qur’an being transcribed from its mother dialect. This can be seen in other places where the use of alif as a vowel is absent yet pronounced, see the following example:
Spelling in Uthman’s [ra] Mushaf (2:9):
It might seem odd that there is a text and you pronounce it differently as to how you read it, but this is the nature of Semitic languages, for referencing, here’s a screenshot of the Al Jazeera News Website (Arabic version, 23 – 09 – 2012), here you will see no vowels, yet when Arabs read the text, they’re able to comprehend and understand it, they’re able speak aloud what the text says even if there are no vowels present:
If it still seems odd, or highly suspicious, here’s a quote from JewFAQ on the matter, “Like most early Semitic alphabetic writing systems, the alefbet has no vowels. People who are fluent in the language do not need vowels to read Hebrew, and most things written in Hebrew in Israel are written without vowels. – ”. Therefore to put this particular claim of ‘textual variants/ orthographical differences’ to rest, it is extremely normal to see this in almost all languages (as shown: Greek, Hebrew, Arabic and English all share this) as they are developing from liturgical transmission to textual transmission and this process is deemed, ‘transcribing‘.
We can also see this in English, with the writing of short vowels, but their pronunciation is that of long vowels:
- Cake is pronounced as ‘caaake’ a long ‘a’ sound, yet one ‘a’ is written.
- Home is pronounced as ‘hooome’, a long, double ‘o’ sound, i.e. oo is pronounced, yet one ‘o’ is written.
For more examples, such as these, see the ‘American English Pronunciation of Long Vowel Sounds‘.
Ordering of Surahs/ Chapters:
Variances in the ordering of Surahs, i.e. Baqarah then Fatihah, do not affect the textual transmission of the Qur’an in the least. The Qur’an is not ordered chronologically, alphabetically, numerically or topically. The order of the Surahs differed in only 3 of 35,000 manuscripts –  and can be due to a myriad of reasons, whoever stacked the papyri into the Masjid’s roof could have been reading those manuscripts and stacked them without replacing them into order, similarly, if a missionary were to open a used stack of playing cards and found a 3 of clubs and then a 6 of hearts directly after, would he claim that the person intentionally placed the cards in that order? There are a hundred reasons why the cards would be in that order, similarly no one should claim to know why in 35,000 manuscripts, only 3 were out of order. Perhaps there was a person who was studying that set, reading that set, or learning to write that set, hence they were put aside and thus taken out of order. Unless the claimant has direct evidence as to why 3 of 35,000 was not in their usual order, they are doing nothing more than making wild guesses.
The existence of the Sana’a Manuscripts do not lend any credence to the claim that the Qur’an is corrupted. Rather the existence of these manuscript – palimpsests, demonstrate the textual integrity of the Qur’an and the grammatico-historical validity of the ahadith which verify the transcribing of the tongue of the Quraysh into a mushaf (text), for which we Muslims still use to this day:
نَّا نَحْنُ نَزَّلْنَا الذِّكْرَ وَإِنَّا لَهُ لَحَافِظُونَ
We have, without doubt, sent down the Message; and We will assuredly guard it (from corruption). – Al Hijr (15:9).
wa Allaahu ‘Alam,
and Allaah knows best.
- Why Muslims Reject the Bible as Scripture.
- The Bible: Inspired Incoherencies [Part 1].
- The Bible: Inspired Incoherencies [Part 2].
- Dr. Gerd Puin and Querying the Koran – Guardian Newspaper [UK].
- Palimpsest – Definition.
- History of the Printing Press.
- Journalism and Mass Communication – Volume 1, History and Development of Mass Communication , Laurie Thomas Lee, Sections 2 “Books” and 2.1 “The Printing Press”.
- Dr. Bruce Metzger and Dr. Bart Ehrman, “The Text of the New Testament: Its Transmission, Corruption, and Restoration”, pp 21 – 22, 4th Ed.
- K. Small & E. Puin, “UNESCO CD of San’a MSS. Part 3: Qur’an Palimpsests, And Unique Qur’an Illustrations”, Manuscripta Orientalia, 2007, Volume 13, Number 2, pp. 63-70.
- Manuscript Table, Islamic Awareness: Sana’a MSS.
- Dr. Bruce Metzger and Dr. Bart Ehrman, “The Text of the New Testament: Its Transmission, Corruption, and Restoration”, pp 21 – 22, 4th Ed.
- Shaykh Mustafa Muhammad al ‘Azami, “The History of the Qur’anic text from Revelation to Compilation”, page 130 – 131.
- Sahih al Bukhari, Book 61: Virtues of the Qur’an, Hadith: 510.
- The Hebrew Alphabet (read: AlifBet), Jew FAQ.
- Does Dr. Gerd Puin’s Research Prove that the Qur’an has been Revised? – Br. Hamza of iERA.